How to Construct Graphs from Data & Interpret Them

What Is a Graph in Science?

In science, a graph is a way of presenting the data collected in an experiment, showing how one variable affects another variable. Experiments are the heart of science; they're how we analyze and understand the world. Scientists refuse to make claims about the world unless they have some hard data, some numbers on which to base that claim. And science experiments are how we get that data.

A science experiment is a way of figuring out the structure and behavior of the world using a systematic method. In any experiment, you change one variable (called the independent variable), and see how it affects another variable (called the dependent variable). Everything else must be kept the same, otherwise it won't be a fair test.

For example, you might want to test how many fruit grow on trees when they're watered by different amounts. The independent variable you're changing is the amount of water the plants get, and the resulting dependent variable you're looking at is how many fruit grow. For this experiment to lead to useful data, everything else must be kept the same. The plants must be in the same kind of soil with the same sunlight. Otherwise, your data would be meaningless because you wouldn't know if the type of soil, amount of sunlight, or other factor was actually causing your result.

Once we have our data, it's time to analyze it. It's time to find the relationship between the two variables.

Constructing Graphs

Graphs are the standard way to present data in science. The most common kind of graph we use to look at the relationship between two variables is called a scatter plot. A scatter plot is where the numbers are plotted on a set of axes by drawing a cross for each of the pieces of data. The independent variable is always plotted on the x-axis, which is the horizontal axis. And the dependent variable is always plotted on the y-axis, which is the vertical axis.

Once you have all your data points, you can draw a line of best fit through the data. A line of best fit isn't actually a straight line. It might be, but it can also be a curve. And a line of best fit doesn't have to go through every data point; in fact, it usually will miss most of the data points. It's just a line that best represents the general shape of the data. Here are a few examples:

Interpreting Graphs

Once we have our scatter plot and line of best fit, it's time to interpret the data, or explain what the data shows.

If the data is completely random and no line of best fit could be drawn, you can say that there is no relationship between the two variables. In our original experiment, for example, this would be like finding that the amount of water made no difference in how many fruit grew on the plant.

If the line of best fit is flat, meaning that it doesn't go uphill or downhill, there is still no relationship between the two variables. This would be what your graph would look like if no matter how much you watered a plant, they all grew exactly the same number of fruit:

If the line of best fit is straight and diagonal, this means that there does appear to be some kind of relationship between the two variables. To be exact, there is a linear relationship. You can also make a statement about how one variable relates to the other.

For example, you might say that the more you water a plant, the more fruit grow, which would be an uphill slope on the graph (a positive linear relationship). Or you might say that the less you water a plant, the more fruit grow, which would be a downhill slope on the graph (a negative linear relationship). Since the graph is a straight line, you can also say that doubling the amount of water you use will always have the same effect on the number of fruit that grow.

Last of all, if the line of best fit is curved, this also means that there is a relationship between the two variables, but that it is a nonlinear relationship. This might mean, for example, that the more you water a plant, the more fruit grow, but once you get to a certain point, it makes less and less of a difference.